Government capex and the order book

Every year, on the first of February, the Finance Minister reads out a number for how much the government plans to spend building things — roads, railway lines, ports, defence equipment, water pipes. The number is large, it is quoted everywhere, and the next morning a cluster of stocks — construction firms, capital-goods makers, cement companies, defence suppliers — often jumps.

The beginner's question is the honest one: if the government is going to spend all that money building things, isn't it obvious those companies will do well? And the answer is the theme of this entire module: yes, the money can reach them — but not as a headline, and not this quarter. It reaches them as an order won, which becomes revenue booked slowly as the thing is built, which becomes cash that arrives later still. Three different clocks, and the share price usually runs on the first one while your money would live on the third.

This module follows one rupee of government capital spending from the Budget page all the way to a company's bank account — and shows you where, along that path, the reader stops and asks the questions that a jumping share price skips.

Of all the ways the government reaches a company, direct spending on building things is the most concrete and the easiest to picture — which is exactly why it is the easiest to over-read.

Start with the word. Capital expenditure, or capex, is spending that leaves a lasting asset behind — a highway, a bridge, a warehouse, a frigate. It is different from revenue expenditure, which is spending consumed within the year and leaves no asset — salaries, interest on old debt, subsidies, pensions. The distinction matters to us because capex is the part of government spending that flows out as orders to companies: someone has to be paid to build the highway. Revenue spending mostly flows to households and lenders, not to a contractor's order book.

There is a further reason capex gets so much attention. A rupee the government spends building a road does not stop at the road. The contractor buys steel and cement, hires workers, and those workers and suppliers spend in turn. Economists call this the fiscal multiplier — how much total activity one rupee of spending eventually stirs up. Capex is thought to have a higher multiplier than most revenue spending, because a durable asset keeps producing value after it is built. So a capex push is genuinely one of the more powerful levers a government has on the real economy.

But "powerful on the economy over years" and "good for this stock this quarter" are two different statements, and the gap between them is where readers get hurt. As the very first part of this shelf kept insisting, the economy is not the market — a real and large flow of government money can be entirely true and still be fully priced into a stock before a single extra brick is laid. This module exists to hold those two things apart.

Follow the rupee. It does not travel in one jump; it passes through four gates, and each gate has its own delay and its own way of leaking.

Gate one — the allocation. The Budget names a figure for capex. This is a plan to spend, an intention for the year, not money already committed to any company. Allocations can be revised down during the year, and actual releases can fall short of the headline when revenues disappoint. So the first thing a careful reader does with a big capex number is treat it as a ceiling and a signal of intent, not a cheque.

Gate two — the award. The money becomes real to a company only when a tender is floated and a contract is awarded. This adds to the winner's order book — the total value of work it has won but not yet built and billed. Awards take months; large ones can take a year from allocation. And they are competitive — a bigger national capex plan can still mean thin margins for the builder if many contractors bid the price down to win the work.

Gate three — execution. Now the thing gets built, over one to four years for a big project. Under the accounting most construction firms use — the percentage-of-completion method, which books revenue in step with how much of the work is done — revenue appears on the income statement gradually as the road is laid, not when the order is won and not when the cash arrives. So a record order book shows up as revenue slowly, spread across years.

Gate four — the payment. The government or its agency pays, often in stages, and often slowly. In the meantime the builder has already paid for steel, cement and labour. The gap between money spent building and money collected is funded out of working capital — the cash tied up running the business — and, very often, out of debt. This is the gate that quietly kills contractors: a full order book, rising revenue, and a bank account bleeding because the customer is a slow payer.

The single most useful habit this chain teaches: when you hear "the government is spending big on infrastructure," ask which gate are we at? The allocation is loud and early; the cash is quiet and late; and the share price has usually already travelled the whole way in its imagination.

Watch the chain run through two composite companies after a large national road-building push is announced. illustrative

Builder A is a road EPC contractor — engineering, procurement and construction — with low debt and a record of collecting its dues. After the push, its order book climbs from ₹8,000 crore to ₹12,000 crore. Because it bid selectively rather than chasing every tender, the new work carries roughly its usual margin. Its balance sheet can fund the two-year build without stretching. Over the next three years, revenue rises steadily as roads are laid, and cash follows within a reasonable lag.

Builder B wins an even bigger book — ₹8,000 crore to ₹14,000 crore — by bidding aggressively to grab share. The extra work carries thinner margins. Worse, to fund labour and materials while payments trickle in, Builder B takes on debt, and its receivables — money billed but not yet collected — stretch out. Two years later its order book still looks magnificent in the headline, but interest costs are eating the profit and the auditors are eyeing the debtor days.

Same trigger. Same "sector tailwind." Opposite outcomes — and the thing that separated them was invisible in the order-book number that both trumpeted. That is the whole lesson of reading a capex cycle: the order book tells you demand arrived; it says nothing about margin, funding, or collection, and those three decide everything.

A big capex allocation is one of the most confidently over-interpreted numbers in Indian markets. Hold on to what it genuinely cannot tell you.

It cannot tell you the timing. The gap between an allocation and a company's cash can run to years, and the market almost always compresses that gap in its head. A stock can rise the day of the Budget and then do nothing for two years while the actual work grinds through tendering and clearances. Being right about the direction and wrong about the timing feels, in your account, exactly like being wrong. As this shelf keeps repeating, precisely when the flow will land is not forecastable — you can trace the channel; you cannot time the payment.

It cannot tell you whether it is already priced. By the time a capex theme is on every front page, the infra and capital-goods stocks have frequently already moved to reflect it. The news is true; the edge is gone. A real tailwind and an already-expensive stock are perfectly able to coexist, and buying the true story at the priced-in price is one of the most common ways to be right and still lose.

It cannot tell you the allocation will actually be spent. Headline capex is a target. Revenue shortfalls, election-year priorities, or slow ministries can leave the money under-spent versus the Budget line. The plan is not the payment.

It cannot tell you which company gets the margin. A larger national spend can still mean thinner margins for builders if competition for tenders is fierce. More demand does not automatically mean more profit per rupee of work — it depends on how hard the bidding gets.

The same handful of errors catch readers around every big capex announcement.

1. Treating the allocation as the spending. The Budget line is intent, not a cheque already written to any company. Between the number and a company's revenue sit tendering, awards, execution and payment.

2. Reading the order book as profit. A record book is promised work, not booked profit. It says nothing about the margin the work carries or whether it will be paid on time — the two things that decide the outcome.

3. Forgetting the working-capital trap. A builder can be drowning in orders and drowning in debt at the same time, because it must fund the build long before the customer pays. Rising revenue with falling free cash is the classic warning.

4. Buying the theme after it is priced. By the time "the capex cycle" is a headline, the stocks have often already moved. The story can be entirely true and the value already gone.

5. Assuming more spend means fatter margins. Fierce competition for government tenders can compress margins even as the national spend rises. Demand up does not equal margin up.